Hacker News Reader

Comment by jawns | Hacker News Reader

A while back, I created an online experiment, based on a 2020 study, that identifies dogmatic thinking.

https://shaungallagher.pressbin.com/blog/dogmatism.html

Without giving too much away, this tracks with the 2023 study's results, in that people with higher intelligence scores are less likely to trust their System 1 gut reaction and more likely to involve their System 2 deliberate, logical processes.

View on HN

Have you read Superforecasting? One conclusion they had was that very partisan thinkers, basically dogmatic, were the worst at predicting future events, and it didn't matter what that dogma was. Liberal or conservative or anti big business or whatever. All of the really good forecasters were relentlessly fact based and spent a huge amount of time diving deeply into facts, and microfacts that supported facts. It sounded exhausting and one of the author's takeaways was "the world is enormously complex" and that most of us don't know anything about anything (hello gelman amnesia).

This is funny because I've heard this from the opposite approach: using forecast accuracy to determine personal model accuracy.

If you can accurately predict the future above random chance, that lends evidence that your internal state aligns with reality.

(There are obviously traps here, eg the conspiracy theorist who sees every outcome as proof of the conspiracy)

> people with higher intelligence scores are less likely to trust their System 1 gut reaction and more likely to involve their System 2 deliberate, logical processes

Amusingly, I mistrust my attention and visual ~memory so much that I reasoned I would find the "easier" task just as hard unless it was literally like 1 vs 3 dots, or unless it stayed up so long that I could count (maybe harder, since I'd likely get more annoyed at myself for missing questions I believed to be easier). I ignored this option because I was fairly sure it'd just mostly waste 200 points.

I went back after completing to check out this option and don't feel like it made the task noticeably easier (to me).

Same. I tried the "easier" version early on in the challenge and it didn't help increase my confidence. I only chose it going forward when I had not even the slightest idea. It really helped once, but not the other times.

That was fun, I scored lower (2.1) dogmatism and higher on the test (840), which they claim supports their hypothesis.

In day to day life the major difference I notice in terms of speed and intelligence personally is that I'm often slow to respond to questions.

It didn't even occur to me to be self-conscious about this until eventually a friend pointed it out, but he kindly framed it as "you think before you speak" and this seems to have served me well.

I had 3.4 and similar number of points as you. The questions don't really work that well.

E.g. the healthcare one doesn't work for my country where universal healthcare is a law and thus State truly has duty to provide it and anyone who believes otherwise is objectively wrong.

I have also received more points for being "dogmatic" in that having strong opinion (that is different than mine of "neutral") on migration is a bad person.

Just because something is the current status quo in your country doesn't mean you can't have a different opinion. I think the question is perfectly valid no matter where you live.

The irony is, faulting the question because "that's just the way it is here" beautifully illustrates the exact premise of the experiment. I'd say that explains a 3.4 score (at least in this game).

It can just be a matter of separating government and society. If you do that, expecting government to follow the law isn't dogmatic at all.

It's also not necessarily dogmatic to believe that you are more informed about an issue than the average person that disagrees with you. Like sure, it can be dogmatic, but it isn't necessarily dogmatic.

I definitely got more dogmatic, but I also find it interesting because I scan visual stuff really fast, and typically notice things a while before people, ie while driving I'll spot a slick-top cop far ahead. So I have a lot of trust in my visual acuity. I got a score of 900 without trying the easy one.

The latter part of the test was 100% incriminating and I think I am quite dogmatic.

> This might be because the least dogmatic participants were more willing to view the easier version of the challenge when they were initially uncertain.

Can’t you test that hypothesis, or do you not have the data?

I also find it super annoying that one of the buttons to choose left is on the right and vice versa.

This link would better serve you and the community as a top-level post. I guess you did share it 2 years ago, though it looks like commenting is now closed. https://news.ycombinator.com/item?id=26620690

Share it again! The iron is hot!

I find the initial paragraphs of analysis rather insistent, one might say dogmatic, in their praise of non-dogmatism. Some later paragraphs express my immediate response: perhaps this dogmatism varies based on your relationship to your current information environment.

I'll suggest that consideration of these numbers might be better supported by tables to show participant and average numbers, and invite comparisons.

And if you share it again, link from (a reply to) your comment.

Neat experiment. Although, I was able to guess a higher than average number of dots correctly, I wonder if I would feel differently had I been under the average.

The explanation given at the end is interestingly to me. However, consider the following alternate: People more likely to answer honestly that they view others opinions as wrong or immoral (ie they would appear to have dogmatic thinking) are also people who consider the easy dot guesses as cheating and desire to avoid cheating. Whereas people who are more likely to want to win the dot guessing by any means are also more likely to lie about their negative views about the opinions of others.

I don't think you can infer that a person answers a 10-minute anonymous online quiz more honestly because they refuse to take a hint out of principle.

It would be a stretch to consider the hints cheating; they are an integral part of the game, they are clearly explained and carry a penalty if used.

I'm not trying to infer it. It's another possibility that would explain results besides the stated one that dogmatic thinking and not looking for more information go hand in hand.

Interestingly enough the dogmatic view is that dogmatic thinking is linked with refusing to seek more information. And here I am trying to be charitable to the "dogmatic” by suggesting that we should look for more information...

The hints felt either like a trick or like cheating to me. That's not a stretch that was just my personal feelings. I didn't even consider that the hints were a game theoretic best way to go if you aren't sure until the end where the explanation hand fed you that exact logic.

I will come back to try this later, but I skimmed it and just wanted to say that the third button gave me a good laugh.

No offense, but the experiment is nonsense.

What makes you think there's any relationship between visual recognition and dogmatism?

Even your own results only show a 2.5% difference between the most dogmatic and least dogmatic participants, assuming they even answered the dogmatic portion of the test truthfully and thoughtfully. I can't see how that amounts to anything more than a statistical error.

Just in case, my score was 940/3.0, although again, I don't see anything to show there should be any relation.

The test states at the end that dogmatism is measured based on the answers to a subset of the questions—the game is simply a correlate.

So, I'm a little frustrated with this because it doesn't tell me how many times I opted for the easier version, just the average, because it seemed like a low cost to pay, I did it almost every time, (which it states is the rational thing to do) but then I also scored as very dogmatic, and got almost exactly average on the test.

The conclusions it drew about this were that I was not open to new information, which is totally off. I DID ask for new information almost every single time, far and above what the average participant requests. The important thing isn't the performance on the test but how often the person requested new information, but then this metric is completely discarded in the final assessment! I'm not sure what exactly this is meant to measure. It feels like it's just smugly calling me dumb for thinking I know better than the average person (which, if I didn't, why would I believe what I believe?)

I only believe things I already have a high degree of confidence in, and it would take good evidence to sway me from those opinions. If this is close minded, being "open minded" seems like a good way for your brain to just fall out of your head.

I scored 1,000 (perfect) on the game and 1.57 on the dogmatism score, which is both lower than the average score and lower than the average in the lowest quartile (people with low dogmatism).

I'm not quite sure what conclusion to take away from this result. While the low dogmatism score is no surprise to me the research results suggest that a low dogmatism-scoring person would be more likely to opt for the "easy version" handicap to verify their hunches but limit themselves to a maximum possible score of 800.

I don't FEEL like I have exceptional perceptual abilities for identifying dots, but I trusted my instincts and honestly rated my confidence as medium (7/10 times) and low confidence (3/10) times.

I'll also add that I'm a little tired and just finished a single beer, so I feel slightly dulled in my faculties right now. I'll probably go to sleep in an hour.

A 2% difference in scores between the most and least dogmatic users means your test complete hogwash.

Your refusal to provide any data beyond averages is very telling. My guess is that there's a very wide distribution of scores, little trend, and very little correlation.

It’s kind of a cool experiment, but I feel like I was hoping for more of a definitive payoff - I just kind of scored as expected, tactically chose the ‘easier’ version if I was less than confident, and at the end just answered along the lines of my usual “nothing is sacred but try to be nice to people” moral framework. Seeing more of a visualization of how my score ranked against others would be cool.

Study link directly, for the curious. (But do the experiment first if you plan to, https://www.pnas.org/doi/10.1073/pnas.2009641117)

Thank you for creating that. I scored 800 and 2,52 in dogmatism.

I thought I would score lower than the average in dogmatism, but didn’t. So I’ll have to revise my belief about my own dogmatism. (:

Fascinating. I scored 800 on the squares and a 2.0 on the dogmatism scale. Thanks for sharing this, the interactive quiz and research is really interesting.

Reply on news.ycombinator.com